Page 1 of 1

M44 repair question

Posted: Wed Aug 01, 2018 4:13 am
by Silverhorse
Hello everybody

for the moment i'm repairing a M44 amp. It has blown powersupply fets and other issues.

There are 4 IRFZ44 fets with their gate resistors of 100 Ohm. If you look closely you can see that there are only 3 resistors of 100 Ohm the most left one measures 51 Ohm. Is this normal? (it looks original)

there is also one diode on location CR13. It's close to cap C32. Does anybody know what the specs are of this diode?

Re: M44 repair question

Posted: Wed Aug 01, 2018 7:50 am
by lilviper
Silverhorse wrote:Hello everybody

for the moment i'm repairing a M44 amp. It has blown powersupply fets and other issues.

There are 4 IRFZ44 fets with their gate resistors of 100 Ohm. If you look closely you can see that there are only 3 resistors of 100 Ohm the most left one measures 51 Ohm. Is this normal? (it looks original)

there is also one diode on location CR13. It's close to cap C32. Does anybody know what the specs are of this diode?
No, and its gotten hot which would probably indicate a failed transisor. The experts will chime in eventually but i see some heat damage there.

Sent from my SM-G960U using Tapatalk

Re: M44 repair question

Posted: Wed Aug 01, 2018 9:22 am
by Silverhorse
there is also an output problem...

there are 4 x A1265 transistors and 4 x C3182 transistors, one A1265 is shorted. What is a good replacement for these?
is it OK to use two different complemantery transistors? Looks like this amp had no repairs before.

Re: M44 repair question

Posted: Wed Aug 01, 2018 10:21 am
by Silverhorse
this is the one that is shorted...

Re: M44 repair question

Posted: Wed Aug 01, 2018 5:13 pm
by lilviper
Im pretty sure there are 2 sets to deal with both halves of the waveform. Pm jacamb2, ericd, ttocs they seem to know what they are talking about.

Sent from my SM-G960U using Tapatalk

Re: M44 repair question

Posted: Thu Aug 02, 2018 3:32 am
by Silverhorse
did some researche

Sanken 2SA1695
Toshiba 2SA1941

Are these OK? Which are the best?
when using these, are there other components to be changed?

Re: M44 repair question

Posted: Thu Aug 02, 2018 3:41 pm
by lilviper
One of these guys has a parts list. Im not sure if its whitenoise or eric.

Sent from my SM-G960U using Tapatalk

Re: M44 repair question

Posted: Fri Aug 03, 2018 6:57 am
by Jacampb2
I use fjl4315/4215 or 2sc5200/2sa1943 pretty much on every PG amp that I repair. Most of the later models used these outputs (they are both the same part, different manufacturers, and are a significant upgrade for the early m series), they should require no modification to use in lace of the originals. You need to replace all the pairs though of you go to these transistors, they are quite a bit different from the originals. The Sanken part you linked above should work fine s well, but unless you specifically want the brand I can't see much reason to pay the premium to get them. You still need to replace all compliments, so at the least the pnp/npn part on the bad channel, but I'd replace them all if you are changing to a different device from stock.

As for the gate resistors, that's not right, see below.

Re: M44 repair question

Posted: Fri Aug 03, 2018 7:03 am
by Jacampb2
The diode in question is a 15v zener, 1n5245b. R21-R24 gate resistors should all be 100R 1/4 flameproof or fusible.

The 51R resistor does look like a factory part, my guess is that someone has seriously fucked up this amp. The 51R resistors are used in the power amp section. Dollars to donuts you will find the wrong resistors in that section because someone clearly swapped one into this gate position.

Good luck,
Jason

Re: M44 repair question

Posted: Sat Aug 04, 2018 6:47 am
by Silverhorse
Thanks Jason for the info.

I've put 4 new gate resistors en 4 new irfz's in, now the power supply works!
The BJT's I need to order.

Do you think that the resistors are switched during the production of this amp? Could this be the reason that this amp failed?
Damn! than I need to find the other one...

Re: M44 repair question

Posted: Sat Aug 04, 2018 10:39 am
by Jacampb2
It would have never passed qc like that. The fet with the lower gate resistor would have been pushed much harder than it's parallel part and it's compliment. Miss matched gate resistors normally cause ps fets to go off like popcorn. If one fails the failure cascades to the others in rapid order. Anything is possible, but I highly doubt it would have passed its birth check let alone 30 + years of use.

My guess is that someone was monkeying with the power amp, pulled some resistors and then forgot where they went...

Good luck,
Jason

Re: M44 repair question

Posted: Sun Sep 09, 2018 7:51 am
by Silverhorse
today I powerd up my M44 .
When I power up, the red LED is working and after a few seconds the green LED is on (the red LED is out). It pulls no more than 3amps.
2 of the 4 channels are working. The left chanel of the front and rear section are working. The right chanel is making noise (sounds very very distorted)

until now I've changed
- the PSU caps
- 4 IRFZ44 fet's
- 4 2SC5200
- 4 2SA1943
- 4 gate resostors

what test could I do now to know what the problem is?

Re: M44 repair question

Posted: Sun Sep 16, 2018 6:07 am
by Silverhorse
because the bad resistor was directly connected to the SG3525 I also replaced it.
I looked after the switched resistor but until now I couldn't find it.
After putting everything together and testing again I'm left with one bad channel. That is the rear right channel. The front left and right are ok.
Through the bad channel I hear the music but it is very distorted. When I raise the volume, the distortion also raises.
When I use the front rca input it has the same result as when I use the rear rca input.

Re: M44 repair question

Posted: Wed Sep 19, 2018 1:25 pm
by Jacampb2
You are going to have to start testing each driver /pre-driver as well as resistors in that channel one at a time. PG used a ton of fusible/flameproof resistors in their poweramps. You almost always find some open. The flameproof resistors are nice in that they don't leave skid marks all over the board when they go, but you have to troubleshoot with a meter because there will be no outward signs that they blew.

Re: M44 repair question

Posted: Wed Sep 19, 2018 1:26 pm
by Jacampb2
Also what changed between your last two posts? First you had two bad channels, then one?

Re: M44 repair question

Posted: Sat Sep 22, 2018 8:41 am
by Silverhorse
Hi Jason

when I changed the SG3525 the problem in the front chanel was solved but not in the rear channel.

it was like you said, a blown 33-ohm resistor that is connected with the part DI 763A.
Can you tell me what the function of this part is? I like to learn and know what I'm doing.

Now the M44 sounds really wonderful. The separation and placement of the voices is really fantastic. Man, these amps are good!

Do I have to adjust the bias of the 4 channels because I changed al de BJT's? What is the procedure to do this?

Re: M44 repair question

Posted: Tue Sep 25, 2018 4:10 pm
by Jacampb2
The top arrow is pointing at a driver transistor and the button one is pointing just above either a pre-driver or we bias transistor.

The open resistor was likely a emitter or base resistor for one of the drivers or pre-driver. When the output shorted they sometimes take up stream components with them.

The SG3525 is the pwm ic for the power supply. It would have zero impact on one channel not working. If it's not working right, the power supply won't be switching and no channels would have worked.

Hopefully you don't have a secondary intermittent problem that temporarily healed itself.

Good luck,
Jason

Re: M44 repair question

Posted: Tue Sep 25, 2018 4:17 pm
by Jacampb2
Oh, yes, bias should be adjusted. Drive the amp for about a half hour until it is at operating temp. Remove all signal and load. Set bias to approximately 2mv dropped across the emitter resistors for each channel. Watch idle current as you adjust, if it goes up over 2a, you did something wrong or you have another issue.

If you have access to an oscilloscope you can also set the bias by driving the amp with a sine wave, place the scope on the output of the 1st channel and adjust bias until the crossover distortion is gone. Continue to monitor idle current while adjusting. Same rule as above, if you go over 2a, something is wrong. Repeat for each channel.

Later,
Jason